Where,
1
r = coefficient reliability between 1
st
half and 2
nd
half X = total number of the 1
st
group Y = total score of 2
nd
group
2
X = square of x
2
Y = square of y Then to know the coefficient of the whole items, the writer used Spearman Brown
Formula:
1 1
1 2
r r
r
k
k
r
= reliability of full test
1
r = reliability of half of the test
The criteria of reliability are : 0.80 - 1.00
= very high 0.60 - 0.79
= high 0.40 - 0.59
= average 0.20 - 0.39
= low 0.00 - 0.19
= very low Hatch and Farhady, 1982:246
3.6.3 Level of Difficulty
To know whether the test items were easy or difficult from the students’ perception who took the test, then the researcher found out the level of difficulty.
To see the level difficulty, the researcher used this formula:
LD =
N R
Where, LD
= Level of difficulty R
= Number of the students who answer correctly N
= Total number of the students
The criteria are : LD 0.30
= Difficult LD = 0.30
– 0.70 = Satisfactory LD 0.70
= Easy Heaton, 1986:178
3.6.4 Discrimination Power
To see the discrimination power see Appendix 10, the writer used the following formula:
DP = the proportion of upper SS – the proportion of lower SS
½ total number students Shohamy, 1985: 81
The criteria are: 1.
If the value is positive, it has discrimination because a large number or more knowledgeable students than poor students get the item correct. If
the value is zero, it means no discrimination. 2.
If the value is negative, it has negative discrimination because more low- level students than high level students get the item correct.
3. In general, the higher discrimination index, the better, in the classroom
situation most items should be higher than 0.20 index. Shohamy, 1985: 82
3.7 Scoring System
In scoring the result s of students’ work, the researcher used Arikunto’s formula
1997: 212. The writer gave score for each correct answer 10 and calculated the student’s score of the pre-test and post-test by using this formula:
Where:
S : The score of the test
R : The right answers
N : The total items
3.8 Data Analysis
In order to know the students’ progress in attempt to master the reading comprehension of narrative text through DRTA strategy, the researcher computed
the students’ score by doing three activities: a.
Scoring the pre-test and the post-test b.
Tabulating the results of the test and calculating the score of the pre-test and
post-test.
c. Data analysis was done by tabulating the result of the test given, that was by
statistically analyzing the data using statistical computerization i.e. repeated measures T-Test of SPSS Statistical Package for Social Science version 16.0
for Windows to see whether or not the difference between pre-test and the post-test was significant, in which the significance was determined by p
0.05. It was used as the data come from the same sample or known as paired data Hatch and Farhady, 1982:114. To find out the significant difference of
students’ reading comprehension of narrative text, the researcher used T-Test, while to find out the significant increase of students’ reading comprehension
100 N
R S
